Perspective-Adaptive Convolutions for Scene Parsing (English)

In: IEEE Transactions on Pattern Analysis and Machine Intelligence   ;  42 ,  4  ;  909-924  ;  2020

How to get this document?

Commercial Copyright fee: €28.50 Basic fee: €4.00 Total price: €32.50
Academic Copyright fee: €28.50 Basic fee: €2.00 Total price: €30.50

Many existing scene parsing methods adopt Convolutional Neural Networks with receptive fields of fixed sizes and shapes, which frequently results in inconsistent predictions of large objects and invisibility of small objects. To tackle this issue, we propose perspective-adaptive convolutions to acquire receptive fields of flexible sizes and shapes during scene parsing. Through adding a new perspective regression layer, we can dynamically infer the position-adaptive perspective coefficient vectors utilized to reshape the convolutional patches. Consequently, the receptive fields can be adjusted automatically according to the various sizes and perspective deformations of the objects in scene images. Our proposed convolutions are differentiable to learn the convolutional parameters and perspective coefficients in an end-to-end way without any extra training supervision of object sizes. Furthermore, considering that the standard convolutions lack contextual information and spatial dependencies, we propose a context adaptive bias to capture both local and global contextual information through average pooling on the local feature patches and global feature maps, followed by flexible attentive summing to the convolutional results. The attentive weights are position-adaptive and context-aware, and can be learned through adding an additional context regression layer. Experiments on Cityscapes and ADE20K datasets well demonstrate the effectiveness of the proposed methods.

Table of contents – Volume 42, Issue 4

Show all volumes and issues

The tables of contents are generated automatically and are based on the data records of the individual contributions available in the index of the TIB portal. The display of the Tables of Contents may therefore be incomplete.

A Hybrid RNN-HMM Approach for Weakly Supervised Temporal Action Segmentation
Kuehne, Hilde / Richard, Alexander / Gall, Juergen | 2020
Automated Video Face Labelling for Films and TV Material
Parkhi, Omkar M. / Rahtu, Esa / Cao, Qiong / Zisserman, Andrew | 2020
Baselines Extraction from Curved Document Images via Slope Fields Recovery
Meng, Gaofeng / Pan, Chunhong / Xiang, Shiming / Wu, Ying | 2020
Deep Self-Evolution Clustering
Chang, Jianlong / Meng, Gaofeng / Wang, Lingfeng / Xiang, Shiming / Pan, Chunhong | 2020
Efficient and Robust Approximate Nearest Neighbor Search Using Hierarchical Navigable Small World Graphs
Malkov, Yu A. / Yashunin, D. A. | 2020
Extracting Geometric Structures in Images with Delaunay Point Processes
Favreau, Jean-Dominique / Lafarge, Florent / Bousseau, Adrien / Auvolat, Alex | 2020
Group Maximum Differentiation Competition: Model Comparison with Few Samples
Ma, Kede / Duanmu, Zhengfang / Wang, Zhou / Wu, Qingbo / Liu, Wentao / Yong, Hongwei / Li, Hongliang / Zhang, Lei | 2020
Hierarchical Bayesian Inverse Lighting of Portraits with a Virtual Light Stage
Shahlaei, Davoud / Blanz, Volker | 2020
Hierarchical Fully Convolutional Network for Joint Atrophy Localization and Alzheimer's Disease Diagnosis Using Structural MRI
Lian, Chunfeng / Liu, Mingxia / Zhang, Jun / Shen, Dinggang | 2020
On Detection of Faint Edges in Noisy Images
Ofir, Nati / Galun, Meirav / Alpert, Sharon / Brandt, Achi / Nadler, Boaz / Basri, Ronen | 2020
Perspective-Adaptive Convolutions for Scene Parsing
Zhang, Rui / Tang, Sheng / Zhang, Yongdong / Li, Jintao / Yan, Shuicheng | 2020
Tensor Robust Principal Component Analysis with a New Tensor Nuclear Norm
Lu, Canyi / Feng, Jiashi / Chen, Yudong / Liu, Wei / Lin, Zhouchen / Yan, Shuicheng | 2020
Tracking-by-Fusion via Gaussian Process Regression Extended to Transfer Learning
Gao, Jin / Wang, Qiang / Xing, Junliang / Ling, Haibin / Hu, Weiming / Maybank, Stephen | 2020
Unsupervised Person Re-Identification by Deep Asymmetric Metric Embedding
Yu, Hong-Xing / Wu, Ancong / Zheng, Wei-Shi | 2020
Visibility Graphs for Image Processing
Iacovacci, Jacopo / Lacasa, Lucas | 2020
Weighted Manifold Alignment using Wave Kernel Signatures for Aligning Medical Image Datasets
Clough, James R. / Balfour, Daniel R. / Cruz, Gastao / Marsden, Paul K. / Prieto, Claudia / Reader, Andrew J. / King, Andrew P. | 2020
Denoising Autoencoders for Overgeneralization in Neural Networks
Spigler, Giacomo | 2020
Efficient Graph Cut Optimization for Full CRFs with Quantized Edges
Veksler, Olga | 2020
Learning Raw Image Reconstruction-Aware Deep Image Compressors
Punnappurath, Abhijith / Brown, Michael S. | 2020
| 2020
Table of Contents
| 2020
| 2020